Real Time Face Tracking

Detect Face Using Haar Cascade
Track Using OpenTLD

FFmpeg Installation

FFmpeg Installation for Windows

1. Open command line prompt on Windows by typing "cmd" at run (Start Menu)
2. At Command Prompt Windows, type cd/ (Change to C:\)
3. mkdir FFmpeg (Make a new folder name FFmpeg at C:\)
4. Extract FFmpeg file to C:\FFmpeg (DL Link : http://ffmpeg.zeranoe.com/builds/)
5. Edit Windows System Path to \bin (Refer to OpenCV tutorial on how to add environment system path)
6. After done, restart command prompt windows by reopen it again 
7. Type "ffmpeg -version" to check whether ffmpeg successfully link
8. Create a "FFmpeg_Example" folder
9. Put all your file inside the folder created
10. cd FFmpeg_Example directory
11. ffmpeg -i....... Start using ffmpeg library

Reference : http://www.wikihow.com/Use-FFmpeg

To change video to image frame

ffmpeg -i video.mpg Pictures%d.jpg

Reference: http://ubuntuforums.org/showthread.php?t=1141293 

Generate list of function dependency in MATLAB

For newer releases of Matlab (eg 2007 or 2008) you could use the built in functions:

1. mlint
2. dependency report and
3. coverage report

Another option is to use Matlab's profiler. The command is profile, it can also be used to track dependencies. To use profile, you could do

>> profile on   % turn profiling on
>> foo;         % entry point to your matlab function or script
>> profile off  % turn profiling off
>> profview     % view the report

If profiler is not available, then perhaps the following two functions are:

1. depfun
2. depdir

For example,

>> deps = depfun('foo');

gives a structure, deps, that contains all the dependencies of foo.m.

Quote from http://stackoverflow.com/questions/95760/how-can-i-generate-a-list-of-function-dependencies-in-matlab

Matlab 2012 link with opencv VS2010 (mex setup)

To use OpenCV library

Download mexopencv library from https://github.com/kyamagu/mexopencv

Prerequiste Components:
VC++ Compiler      (included in VS2010)
Microsoft SDK 7.1 (included in VS2010)
Matlab

1. Open Matlab, type mex -setup in command windows
2. Check whether there is any default compiler
3. If no default compiler, choose a compiler from the list
4. Type command cv.make('opencv_path','your\opencv\path')- replace your opencv path with your opencv directory (E.g. 'C:\OpenCV2.4')
5. Let it compile
6. After compiled, type addpath('mexopencv_path') - This must be called first for every project that used opencv library
7. Type 'help cv' to display list of opencv command
8. Can start using opencv library

Tutorial

http://www.cs.stonybrook.edu/~kyamagu/mexopencv/

https://github.com/kyamagu/mexopencv/blob/master/README.markdown

List of opencv function

http://www.cs.stonybrook.edu/~kyamagu/mexopencv/matlab/

OpenCV 2.4.2 Visual Studio 2010 (VS2010) Windows 7 64bit setup

Installation of OpenCV 2.4.2 for Visual Studio 2010 (Windows 7 x64)

Download the latest OpenCV library from sourceforge (2.4.2 or 2.4.3)

1. Right Click My Computer -> Properties -> Advance System Setting -> Environment Variables -> Path -> Type in OpenCV dir ("C:\OpenCV2.4\build\x64\bin")
2. Open Visual Studio 2010. Click New Project -> Select Win32 Console Application -> Check Empty Project -> Finish
3. Click Property Manager Tab (Bottom right) -> Create 2 New Property Sheets with preference name ( OPENCV_DEBUG & OPENCV_RELEASE)
4. Open OPENCV_DEBUG Property Sheet
1. For C/C++ - Add path for Additional Include Directories (OpenCV2.4\build\include)
2. For Linker/General - Add path for Additional Library Directories (OpenCV2.4\build\x64\vc10\lib)
3. For Linker/General - Add the following lib file (For OpenCV2.4.3). The list of lib file can be found in OpenCV2.4\build\x64\vc10\lib:
   
   opencv_core243d.lib
   opencv_imgproc243d.lib
   opencv_highgui243d.lib
   opencv_ml243d.lib
   opencv_video243d.lib
   opencv_features2d243d.lib
   opencv_calib3d243d.lib
   opencv_objdetect243d.lib
   
4. Repeat the steps for OPENCV_RELEASE, and change the lib file for Linker/General to:
   
   opencv_core243.lib
   opencv_imgproc243.lib
   opencv_highgui243.lib
   opencv_ml243.lib
   opencv_video243.lib
   opencv_features2d243.lib
   opencv_calib3d243.lib
   opencv_objdetect243.lib
   
   same filenames with 'd' - debug version removed
5. To run x64 environment
   Go Build -> Configuration Manager -> Active Solution Platform -> New -> Change Itanium to x64
6. Compile and run opencv code
7. If tbb_debug.dll not found (for OpenCV 2.4.2), (OpenCV2.4.3 does not has this problem). Go download the tbb_debug.dll file from Intel Threading Building Blocks. extract x64 version and place into OpenCV\build\x64\bin
8. if 32-bit win, extract x86 version out and place into OpenCV\build\x86\bin
9. Compile your opencv code

Note: 

For some PC there will be Illegal Instruction Error when running cvHaarDetectObjects library (2.4.3 version). (Rare case). 

Solution: Use 2.4.2 version

Hello World!